FACTUAL INFORMATION
History of the flight
The Robinson Helicopter Company Model 22 helicopter (R22), registered VH-OHA, was being operated to conduct dual training in the Bankstown training area. An experienced flight instructor was conducting a student pilot’s second flight lesson. The planned lesson was to demonstrate and practice climbing and descending manoeuvres. The weather in the area was reported as no cloud and light wind. This description was consistent with the forecast conditions.
A witness located to the south of the helicopter’s flight path reported observing the helicopter. This witness reported that there was a sound, similar to a car backfiring, followed almost immediately by a louder bang. The witness reported observing what appeared to be a main rotor blade separating from the helicopter before losing sight of the helicopter as it descended behind trees.
The helicopter impacted the ground in an inverted attitude. Both occupants received fatal injuries.
Wreckage information
The wreckage trail of the helicopter commenced on flat ground to the west of Bringelly Creek, a tributary of the Nepean River, and extended along a bearing of approximately 240 degrees. The wreckage trail continued up an escarpment and ended on flat ground on the top of the escarpment. One of the main rotor blades could not be located during the initial examination of the wreckage trail and the accident site. The blade was located on the following day after a wider search.
The length of the wreckage trail was approximately 700 m, with the cabin and engine located in the most westerly section of the wreckage trail. The separated main rotor blade was located approximately 135 m north of the wreckage trail centreline. The other main rotor blade, which remained attached to the main rotor head, was located approximately 55 m to the east of the main cabin and engine.
Personnel
The flight instructor held a valid commercial pilot’s licence (helicopter), a grade one flight instructor rating, and endorsement for the R22 helicopter. He had recorded in excess of 3,800 hours total time in helicopters, with in excess of 2,700 hours in the R22 type. He held a valid class one medical certificate. He was reported to be fit and well rested and was flying within required flight and duty time limitations.
The student was conducting his second flight lesson and held a valid class one medical certificate. He was reported to be fit and well rested prior to commencing the flight.
The helicopter
The R22 helicopter was manufactured in the USA, in 1991, as a Robinson Helicopter Company Model 22 Mariner. It was exported to Australia and placed on the Australian Civil Aircraft Register in December 1996. It was operated privately when first operated in Australia. It was subsequently sold in 1997 and was operated by a flying school for a period of approximately 1 year. In 1998, it was sold to a helicopter sales company where it stood idle for a period of approximately 6 weeks, before it was purchased by an aerial mustering company and flown to the Northern Territory for use in aerial survey, as well as cattle mustering operations. In 1999, it was purchased by the current owner.
The total time in service (TTIS) of the helicopter was recorded as 1986.2 hours at the time of the accident. The investigation to date has revealed that there were some minor arithmetic errors in the recording of the time in service. The TTIS of the helicopter has been checked and recalculated as 2009.5 hours. However, the investigation into establishing the history and verifying the operating hours of the helicopter is continuing.
Examination of the failed main rotor blade
The failed main rotor blade was examined in the ATSB laboratories and exhibited features consistent with a fatigue failure of the main rotor blade root fitting, in the counterbore of the bolt hole in the area of rotor station 10.35 (inboard bolt hole). It is the same site that has been identified by the manufacturer as being the critical location for fatigue cracking during fatigue testing.
Microscopic examination of the failed surface revealed evidence of corrosion at the site of the fatigue crack initiation point. Significant areas of breakdown of the adhesive bonding between the rotor blade skin, the end of the spar, and the rotor blade root fitting were also observed. There was no evidence of cracking on either the upper or lower rotor blade skin in the area adjacent to the fatigue crack. Examination of the other blade from the accident helicopter revealed similar disbonding of the rotor blade skin end of the spar, although to a lesser extent.
Six other R22 main rotor blades were obtained by the ATSB for the purposes of examining the adhesion between the end of the spar, the root fitting and upper and lower skins. These other blades exhibited similar disbonding, to varying extents, of the adhesive in the area of the end of the spar and the rotor blade root fitting. One example extended past the first and second bolt holes in the blade root fitting.
Incomplete adhesive filling, in the form of an elongated void of the gap that exists between the edge of the spar, the root fitting and the skin, was also observed in two blades.
The ATSB issued recommendation R20030186 on 17 September 2003. This recommendation stated: `The Australian Transport Safety Bureau recommends that the United States Federal Aviation Administration, in conjunction with the manufacturer of the helicopter, the Robinson Helicopter Company, conduct a review of a representative sample of main rotor blade root fittings to establish the integrity of the adhesive bond in the spar to root fitting joint. The review should establish the extent of the loss of adhesion and the extent to which corrosion has infiltrated in the region of the inboard bolt hole of the blade root fitting. If possible, where disbonding is discovered, the operating history and in-service flight spectrum of the helicopter and the environmental conditions under which it operated should also be assessed.’
Initial findings of the examination of a sample of blades by the manufacturer revealed that these other blades exhibited similar disbonding, to varying extents, of the adhesive in the area of the end of the spar and the rotor blade root fitting to rotor blade skin attachment.
The initial response from the Federal Aviation Administration (FAA) was the issuing of a Special Airworthiness Information Bulletin (SAIB) SW-04-36 on 17 December 2003, that reiterated already existing information. The formal response to the recommendation from the FAA did not expand on the information contained in the SAIB and indicated that they believed the blade had failed from under-recording of time in service.
The ATSB subsequently wrote to the FAA asking them to reconsider their response with regard to the evidence that was presented to them and with reference to an overseas accident (see below). The FAA responded by issuing FAA Emergency Airworthiness Directive (AD) 2004-06-52. The Australian Civil Aviation Safety Authority responded to this action from the FAA by issuing AD/R22/31 amendment 9 and subsequently AD/R22/31 amendment 10. These ADs required, among other things, that all operators retire R22 main rotor blades at 10 years calendar life, to conduct a tracking and balancing of the main rotor blades and to replace any main rotor blade that developed vibration within 5 hours following the tracking and balancing of the blades.
Ongoing investigation aspects
The ATSB is continuing to assess the failure mode of the adhesive bonding in the main rotor blade-to-root fitting on Robinson Helicopter Company R22 helicopters. The Bureau is also continuing to establish the operating history and time in service of the accident helicopter.
Similar accident
On 29 February 2004, a R22 helicopter registered and operated in Israel crashed following the loss of a main rotor blade in flight. Both occupants received fatal injuries. Following this accident the ATSB established contact with the Israeli Ministry of Transport who were responsible for the accident investigation.
Information provided by the Israeli investigation team on the examination of the blade revealed that it had also failed as a result of corrosion initiated fatigue within the blade root. The total time in service of the failed blade was approximately 1490 hours.
The investigation is continuing.....
source